# Solid State Thyratron using SCR's?

#### onereddog

Joined Jan 12, 2010
2
Hi all,

New member, first post so here goes!

Im looking for some tips & hints to replace an old hydrogen thyratron with a solid state device using thyristors. The application is a small radar, PRF about 800Hz & pulse width 1uS. I intend to run a bank of say 4 SCRs in series to switch a 3kV charge voltage to a pulse package. 2 x 1:1:1 gate pulse transformers for the triggering.

Has anyone seen this done before?

Pitfalls? Do's & dont's?

Cheers..

#### SgtWookie

Joined Jul 17, 2007
22,221
Interesting; I'd never heard of a hydrogen thyratron before and had to look it up! I thought magnetrons and klystrons were "old school"; that goes back a good bit further. TWT's took the place of maggies and KPA's for airborne stuff a good while ago.

Since you're working on that sort of thing, I'll assume that you have a good bit more than a passing acquaintance with electronics. Unfortunately, this site is primarily geared towards complete "n00bs", and the type of equipment you're working on involves very lethal power levels. We don't want our n00bs to get the idea that it's OK to experiment with this sort of thing casually; as a high-power radar can kill you even at a distance.

But just for argument's sake; once you switch on an SCR or TRIAC, it won't turn off until the current flow nears zero. This would give a very long and decaying output pulse.

You're going to need more like an RF transistor or high-frequency IGBT (I don't even know if they make the latter for high frequency stuff yet).

#### beenthere

Joined Apr 20, 2004
15,819
Scr's have a sensitivity to ΔV/ΔT. If the voltage rises too fast, the device becomes a dead short, and eats the gate drive circuitry.

You're either stuck with finding a source for thyratrons or discovering some work-around. They still make radars, so there must be something. And the next person posting may have the answer...

#### BillB3857

Joined Feb 28, 2009
2,547
But just for argument's sake; once you switch on an SCR or TRIAC, it won't turn off until the current flow nears zero. This would give a very long and decaying output pulse.
Thyratrons do the same thing,Sgt. Once triggered, they keep conducting through the ionized gas until current is reduced below holding levels. Usually, this is the result of the PFN (Pulse Forming Network) which actually shapes the high voltage pulse to the magnetron. The PFN consisted of a network of inductors and capacitors. Charged from a high voltage source, it releases all the stored energy as a shaped pulse through the thyratron to the maggie. Various networks were used in order to provide an appropriate pulse width, depending upon the mission of the radar. Search radar had fairly wide pulse width to translate into more power. Narrower pulse widths were used for things such as fire control (no, not putting out a fire but aiming armament). Tug boats also use a very high PRF (Pulse Repetition Frequency) along with a very short pulse with in order to allow close range detail to be observed.

Thyratrons were also used industrially in things such as welders and variable speed controls, just as SCR's are today.

#### SgtWookie

Joined Jul 17, 2007
22,221
BillB3857,

I'm familiar with fire control radar; I was MOS 6657. Our F-4J and F-4S Phantom II's had 1,525W main KPAs and 900W CW illuminator KPAs, and I frankly forgot about the tubes in the LRU-2A3 until a few minutes ago. Transmitter had 10.5KV and 17KV sections. The fire control radar had Short Pulse (0.65uS pulse) "Chirp", and PD modes. They may have touched on thyratrons in school, but that was many moons ago. KPA's were obsolete even by the time they were new to me in the early 70's. The AWG-9 in the F-14's had 10KW TWT's in them.

#### BillB3857

Joined Feb 28, 2009
2,547
Early '70s. My experience was with APS-20 and APS-45 as well as the navigation and communications gear, antenna stabilization, ECM, IFF, etc on my avitar.. Got out in Nov of '64. Went to work for McDonnell in the Numerical Control arena (now known as CNC)

#### SgtWookie

Joined Jul 17, 2007
22,221
I worked on the AN/AWG-10 and -10A's, which included the AN/APG-59 radar (Westinghouse design). It was developed before IC's, so Westinghouse made their own IC's out of discrete components, and cast them in epoxy. The IMA guys had to have an incredible assortment of these modules sitting around to fix stuff. By the time I was in, our airplanes had the stuffings beat out of 'em from thousands of hours of flight time; keeping 'em up was a major ordeal. We had some built-in test functions, but they were pretty rudimentary. The newer systems are a piece of cake to maintain by comparison.

Last edited:

#### beenthere

Joined Apr 20, 2004
15,819
As long as we're off topic, Hughes (we called 'em Huge Aircraft) did a brisk business in encapsulated modules. One having no more inside than a transistor and three resistors set Uncle Sugar back about $40 in the 60's. IBM did some of that too. Unique IC's will same old TTL inside. #### BillB3857 Joined Feb 28, 2009 2,547 #### k7elp60 Joined Nov 4, 2008 562 I remember the thyratrons as well that were used to discharge pulse forming networks for mostly radar. I just realized that the old relaxtion oscillator with a neon bulb in parallel with a capacitor and a resistor in series to a dc voltage is very similar. The capacitor charges up until the neon gas ionizes, causes the capacitor to discharge until the neon lamp goes out and the whole process repeats. I can think of a modern transmitter(close to Megawatt pulse power) that uses SCR's to discharge a pulse forming network. It is the Loran-C transmitters. #### BillB3857 Joined Feb 28, 2009 2,547 If my calculations are correct, the max range of the OPs radar is right at 100 miles, based upon PRF and 12.36 microseconds / radar range mile. #### SgtWookie Joined Jul 17, 2007 22,221 100 miles sounds about right. Hughes had a pretty good thing going on with the F14's. The delivered aircraft were$30mil/each. The radar system alone was \$14.7mil each for the F-14A, right up until the F14D program. The Navy finally balked, and threatened to kill the program. Many of the engineers were put out the door. The assembly line was moved from the El Segundo facility to Mexicali, MX; the supervisors and managers lived in Calexico, CA and commuted daily across the border.

The mid-90's was not a happy time to be in Aerospace & Defense.

#### BillB3857

Joined Feb 28, 2009
2,547
Tell me about the late '80s early '90s. That is the time frame McDonnell Douglas lost both the A-12 and the YF-23. So many good people let go. Really heart breaking to have to be one of many handing out the paperwork.

#### SgtWookie

Joined Jul 17, 2007
22,221
Yeah, it was a really heartbreaking time. 2/3 of the people at Hughes got put out the door from '87 to '94.

#### 3ldon

Joined Jan 9, 2010
82
I think it would be much less effort to wire up 4 IGBT's in series, than attempt to find a SCR that can turn on in less than 10usec.

#### onereddog

Joined Jan 12, 2010
2
hi all,

Thanks for your replys and yes, this is dangerous stuff so suitable experience essestial. While Magnetrons & Klystrons still have active roles these days Thyratrons are pretty much obsolete as IGBT's can be used with suitable design PFN's to fire a Mag.

I recall a Plessey radar that used SCR's in series to simulate a thyratron (the kick back emf pulse from the PFN provided the switch off potential) Snubbing networks were used to cure dv/dt issues. I also recall the transmitter had the most problems!

Anyway I'm going to give it a go. May even consider IGBT's as suggested.

cheers..